Logical Philosophy
by Andrew Boucher
v1.00 Last updated: 1 Dec 2000
Please send your comments to abo

Note to the reader: To avoid confusion and possible misinterpretations of the author's intentions, whenever a paragraph contains a definition or explication of how the author means the meaning of a word, asterisks have been placed after the paragraph number and before the word or words in question. The reader is warned that some words may be meant idiosyncratically.


1.1 I mean meanings of words and sequences of words.

1.2* What is meant by a word (sequence of words) is the *meaning of that word (sequence of words).

1.3* A *meaning of a word (sequence of words) is intermediate between the word and the things themselves. (From now on, I will drop the parenthetical 'sequence of words.')

1.4 I do not have the meanings of words in hand. I cannot hear, see, or taste meanings of words. I know meanings of words only in so far as my intuitive feeling for meanings of words goes.

1.5 One does not translate meanings of words into words, nor words into meanings of words. Meanings of words go side by side with the words.

1.6 One does not mean meanings of words by means of mental pictures or images. One means by meaning (meanings of words are what they are).

1.7 (Sometimes, it does seem that, when one concentrates very hard in response to a question of what one means by some word, one does conjure up a picture. But this does not entail anything other than what it says (or less); in particular, it does not entail that one means meanings of words using mental pictures or images.)

1.8 Although meanings of words themselves are (of course) absolute, what particular words mean may change.

1.9 Different individuals may mean the same word quite differently: for example, 'God'.

1.10 Indeed, the same individual may mean the same word differently at different times: a kite may either be a type of bird or a type of toy.

1.11 We may never, by the same word, mean exactly the same thing over time, or we may, by some word, always mean the same thing (or, somewhere in between). It does not seem that I am able to remember how I meant a particular word, so I cannot tell by introspection. (At most I remember the circumstances of my use of the word.) However, we seem to assume constancy of meaning (which, over long time periods, is probably wrong for many words).

1.12 On the other hand, two words may often be meant in roughly the same way. Consider 'pail' or 'bucket,' or 'young male human being' and 'male youth.' (Again, I cannot positively assert that I do mean the different words in exactly the same way.)

1.13 The connection between words and meanings of words is logically arbitrary (but causally quite complex).

1.14 It is possible that any word have any meaning of a word.

1.15 A meaning of a word is not in that word: words do not mean by themselves, but through me (or, presumably, someone else).

1.16 "The meaning of word w" thus must have an implicit indexical as to who is meaning and when he is meaning it.

1.17 Often, the indexical seems to be me now, or to the English-speaking community at the current period of time.

1.18* A *meaning is a meaning of a word.

2.1 'Meaning' has two principal meanings (under both meanings!). On the one hand, it means what is meant. On the other, it means the extension of what is meant, what the word sequence is supposed to symbolize.

2.2 Meanings, unlike extensions in some cases, are never tangible. Nonetheless, one is naturally led to conclude that "extension" is not the only meaning of 'meaning.' For, we would say that 'giraffe' has a meaning, even though there turned out to be no giraffes. Again, 'Pegasus' is meaningful, even though, presumably, Pegasus does not exist.

2.3 In fact extensional meanings depend on intensional meanings. A word symbolizes an extensional meaning only because we (intensionally) mean by it; e.g. 'red' symbolizes a red patch because we mean 'red' the way we do.

2.4 One must be careful in posing and answering the question, "What is the meaning of 'meaning?' " Here, the answer will affect the question: and thus one may provide an answer which happens to be internally consistent, but which changes the question originally asked.

2.5 "The meaning of a word is its use" is an example of an incorrect answer to the original question. However, no knock-down argument can be given for this, in the same way that no one can argue that red is not blue.

2.6 Admittedly, it does seem that language is used as well as meant. And this merely says: we sometimes use words, just seeking a reaction, or an outlet to feelings, etc. rather than, or as well as, mean them. It does not imply that we do not mean meanings of words.

2.7 Indeed, there may be some words which are only used and not meant. But this does not imply the same is true of all words. And it is not true of all words.

2.8 Some philosophers equate, more or less, the meaning of a word with how we learn it. But surely it is possible that we never learned the meaning of any particular word (or, for that matter, never previously used it, or had to explain its meaning), since it is possible that the world just began its existence one minute ago. And so any connection between the two is not a necessary one (it seems to be merely causal).

2.9 "Meanings do not exist" is either unimportant or wrong. If the gist is that meanings are not physical objects or attributes or places or times or etc., then the assertion is unimportant, and I agree with it; meanings are meanings of words. However, if the gist is that I do not mean meanings of words, then one is wrong, for I do. Similarly, claiming that 'meaning' does not have any meaning is wrong: for, similarly, it does. And claiming that I cannot or do not refer to meanings is wrong, for I can and do refer to them: e.g. "the meaning of 'red' " refers.

2.10 I was once asked whether meanings are "over and above" and "independent" of physical objects. In one sense of 'over and above,' clearly they are not: for it is a physical object that means or could mean the meaning of a word. In another, clearly they are: for a meaning is not some combination of physical objects.

2.11 ("Even if there were no human beings, there would be meanings" is not the case if the sentence in quotes means "There would be people meaning meanings of words even if there were no people," while it is the case if what it means is "There could be people meaning meanings of words even if there were no people.")

2.12 Some philosophers do not deny meanings, but rather claim they do not need to talk about them. Of course, it is not necessary to talk about any subject. Indeed, one can live without thinking about any philosophical questions. However, if the claim is that "meaning" is never useful, then it should be noted that "meaning" is very often used in this treatise.

3.1 A meaning may not be precise, in the sense that it clearly determines what falls under it; rather, the meanings of most words seem to be blurred. (The evidence for this is an intuitive uncertainty when presented with certain situations and sentences or questions using the words. Also, for many words, it would seem very strange if there were a sharp boundary in each of their meanings.) Usually, the blur seems small relative to the number of situations which we would say are definite. (Here, think of meanings as various degrees of shading on a plane, where only a small area is grey, while the rest is black or white.) For example, while "rain" seems blurred, most situations (thunderstorms, clear days) definitely either fall under or do not fall under it, while only a few (one raindrop at eye level) cause trouble.

3.2 The meanings of some words do not seem to be blurred, among which I would include "one."

4.1* *Concepts are meanings.

4.2* A *concept is a sort of meaning: an integral meaning (a meaning which hangs together in one piece), a meaning which is (as others have called what I intend) "logically possible." Thus, all concepts are meanings, but not all meanings are concepts.

4.3 ("Yellow and red (together)," "moving faster than self," "fast to me and slow to me," "square circles," "The circle is square," and "the horse the dog" are all not concepts. "Seventeen yellow pigs," "seeing oneself," "well and slow," "fast to me and slow to you," "the small circle," and "the circle is small" are all concepts.)

4.4* A *concept is neither long nor short, complex nor simple (unless one wants to say that all concepts are simple). Word strings are long or short, but any concept may possibly be meant by a single word. Concepts are not complex, but some concepts are more detailed than others: e.g. "a black-haired boy" is more detailed than "a boy."

4.5 There is not (and cannot be) some algorithm which determines whether a sequence of words means a concept or not. Usually all that one can do is consider the meanings of the words and the grammar of the sequence and see what one's intuition says about whether some meanings conflict, whether the combination of meanings fuses into an integral meaning. We do not "know" meanings other than by our meaning of them. Meanings are not in hand and manipulable. We cannot place one over another to see if they conflict. So, in a large number of cases, we will not be able to tell, by examination, whether a sequence means a concept. (One need only consider mathematical assertions to see this is so.)

4.6 Indeed, while it may seem reasonable to assert that every meaningful individual word means a concept, there does not seem to be anything necessary in this, and certain words like 'God' are suspect.

5.1* A *referring concept tries or has the potential to refer. ("Snow," "George Washington," "the tallest man," "this procedure," and "he" are all referring concepts.)

5.2 What a referring concept tries to refer to may depend on its context. ("Snow" is context-independent; "the next word" is context-dependent.)

5.3 "Kite" is context-independent. Although 'kite' may be meant in one of two ways (to signify either a bird or a toy), this happens not by it having one context-dependent meaning, but two context-independent meanings.

5.4* The *referent or *referents of a referring concept in a particular context are what the concept refers to in that context. The referent of "George Washington" is George Washington (the President, since that is how I mean 'George Washington'). The referent of "my next sentence," in the context of being spoken by me at a particular time, is the next sentence I will utter. The referents of "the chairs in this room" in the context of my being in McConnell 406 are the chairs in McConnell 406.

5.5 There are many natural sorts of referring concepts, among which are: proper names ("George Washington"), names ("running" and "wisdom"), descriptions ("the tallest man in the room"), pointers ("these books"), and alluding referring concepts ("the former" and "they"). These divisions are not always sharp; for instance, "those nine men" is both a description and an alluding referring concept.

5.6* "Some cows in the field" is not a *referring concept. The idea is: a referring concept must try to refer to definite things.

5.7 Some word sequences look like they would mean a referring concept, because they have a structure similar to a word sequence which does, but in fact they do not. For example, 'square circles' looks like it would mean a referring concept, but it does not, because it's not the case that "square circles" is a concept.

5.8 Some reference, even though attempted by concepts, still fails: what the reference purports or pre-supposes is not there as purported or pre-supposed.

5.9 Examples of failing referring concepts are: "the King of France," "when Cicero was emperor," "Becky's writing to me," when presumably (and annoyingly) Becky has never written to me, and "the sister of the King of France."

5.10 A referring concept's reference may depend on another referring concept's reference. For instance, since "the referent of the next referring concept" precedes "George Washington", then the reference of the former depends on the reference of the latter, and so they both refer to George Washington.

5.11 It is possible to construct examples of failing reference due to vicious or unending dependence, including:

[A] "the referent of the referring concept 5.11A"
"the referent of the next referring concept"
"the referent of the preceding referring concept"
"the referent of the next referring concept"
"the referent of the next referring concept"
"the referent of the next referring concept"

5.12 Note that referring concepts may refer to themselves, e.g.

[A] "the referring concept 5.12A"

5.13 A referring concept is an arrow. (It fails when it cannot reach its target.)

6.1* An *assertion (a *statement) is a meaning.

6.2* An *assertion is purely assertoric, i.e. it must not contain literary devices or asides or the like.

6.3 Not all assertions are concepts, e.g. "My square circle is red" is an assertion which is not a concept.

6.4 An assertion which is a concept is a *proposition. So, a proposition is a concept which asserts. ("Toby was a dog" is a proposition.)

6.5 A proposition tries to assert something about the way the world is. A proposition describes.

6.6 What a proposition asserts about the world may depend on the context of the proposition. ("2 + 2 = 4" is context-independent. "Yesterday the weather was beautiful" is context-dependent.)

7.1* An *atomic statement tries to assert of some thing or things, and asserts something of them. ("Snow is white" asserts of snow that it is white. "Most Americans speak English or Spanish" asserts of Americans that most of them speak English or Spanish.)

7.2 Distinct atomic statements do not necessarily make independent or disjoint assertions. E.g. "His body is green" and "His fingers are blue" are both atomic but not independent.

7.3 It may not be definite of what an atomic statement which contains two or more referring concepts is asserting. For example, "Mario telephoned Tom" could either be considered to assert of Mario, of Tom, or of both Mario and Tom. And so "Mario telephoned Tom" could, respectively, be asserting the being telephoned by Mario of Tom, the having telephoned Tom of Mario, or that the first telephoned the second of Mario and Tom. Although we shall not do so, this indefiniteness could be eliminated by stipulating which choice to make. (An example of an atomic statement which contains two referring concepts which is not indefinite would be "Mario and Tom are tall," but this plays on the fact that the subject is one referring concept which is the conjunction of two referring concepts.)

7.4 A *simple statement contains one referring concept (perhaps containing other referring concepts) which attempts to refer to a thing or things, and it asserts a property that the thing or things are supposed to possess. ("Snow is white" and "Mario and Tom are tall" are simple.)

7.5 All simple statements are atomic, but some atomic statements are not simple. ("Mario telephoned Tom" is atomic but not simple, containing two referring concepts.)

7.6 A metaphor for atomic statements is: a picture with arrows. A metaphor for a simple statement is: a picture with an arrow. In the case of a simple statement, what the picture and the arrow are, should be fairly evident. In the case of atomic statements, some analysis may be needed. ("Snow is white" can be represented as "a white thing: snow." "Mario telephoned Tom" as, among others, "one telephoned another: Mario, Tom.")

8.1* A statement may embody atomic statements, due to logical connectives. Such a statement will be called *molecular. ("Molly plays the flute and Marco has curly hair" is a molecular statement, which embodies two atomic statements.)

8.2 Statements can only be embodied in molecular statements. A statement which refers to another statement S does not embody S, but rather embodies a referring concept which refers to S.

8.3 ""Northampton is peaceful" is a proposition about a small city" embodies not the proposition "Northampton is peaceful," but instead the referring concept ""Northampton is peaceful"", which refers to the proposition. The proposition and the referring concept are different, although this is hard to see from the notation we use to refer to them.

8.4 In any case, "I believe that this is a pen" does not even embody a referring concept which refers to the proposition "This is a pen." (Nor is there any reason to suppose that what one believes is a meaning, so a statement or a proposition, i.e. that what one believes is identical to what one means.)

8.5 A molecular proposition may embody an atomic statement which is not a proposition. ("The square is a circle or Carrie runs fast" is a molecular proposition, even though the first embodied atomic statement is not a proposition.) Whether this can happen, and when, of course depends on the logical connectives embodied in the proposition. (For instance, if "and" connects two atomic statements, then both atomic statements must be propositions for the molecular statement to be a proposition.)

9.1* The denial of an assertion is an *assertion.

9.2 A denial is a molecular statement.

9.3 To deny an atomic statement one does not take the negation of the picture and use the same arrows; one simply rejects the particular picture with arrows.

9.4 The denial of "The King of France is bald" ("Bald: the King of France") is not "The King of France is not bald" ("Not bald: the King of France"), but "It is not the case that the King of France is bald" (Not: bald: the King of France").

9.5 Nonetheless, when there is no referential failure, the denial of an atomic statement is equivalent to the negation of the picture with the same arrows. Because Dmitri exists, the denial of "Dmitri has curly hair" ("Not: has curly hair: Dmitri") is equivalent to "Dmitri does not have curly hair" ("Does not have curly hair: Dmitri").

9.6 Confusion about denial can lead to philosophical difficulties, such as the following fallacious proof of the existence of God: "God does not exist" is self-contradictory; but it is the negation of "God exists;" hence God exists. Indeed, "God does not exist" is self-contradictory, but the negation of "God exists" is not "God does not exist" but "It is not the case that God exists". And "God does not exist" and "It is not the case that God exists" are equivalent only in the case when "God" does not fail to refer, i.e. only when God exists. (Of course, people speak loosely, and so often use 'God does not exist' to mean "It is not the case that God exists" and not "God does not exist.")

9.7 Either: a statement, or its denial.

9.8 Either a statement, or its denial, even in blurred cases, for example "it is raining" when there is one raindrop at eye level. For if we admit that we cannot quite say it is raining in the situation of one raindrop, then we deny it is raining, i.e. it is not raining.

10.1* A statement is *true in a particular context, if what it asserts in that context holds. A context-independent statement is *true, if what it asserts holds. ("Yesterday the weather was beautiful" is true when asserted in the context of a day immediately following a beautiful day.)

10.2 All true statements must be propositions. (A truth must be logically possible.)

10.3 Returning to our metaphor, an atomic proposition is true when the picture fits the referents of the arrows.

10.4 Whether what a proposition asserts holds may depend on whether what another proposition asserts holds. For instance, since "The next proposition is true" precedes "George Washington was America's first President", whether the former is true depends on whether the latter is true.

10.5 It is possible to construct examples of vicious or unending dependence, including:

[A] "The statement 10.5A is true"
[B] "The statement 10.5B is not true"
"The next statement is true"
"The previous statement is not true"
"The next statement is not true"
"The next statement is not true"
"The next statement is not true"
All these statements are not true, because "what they assert" suffers from referential failure, so it cannot be the case that what they assert holds.

10.6 It is therefore possible both that S and that S is not true. For instance, the statement 10.5A is not true and the statement 10.5A is not true. ("S" and "S is not true" are here in fact the same proposition.)

10.7 Another example: All statements are true or not true, but "All statements are true or not true" is not true.

10.8 And another example:

[A] "The statement 10.8A is true or snow is white"
is not true, even though "snow is white" is true. Of course, because snow is white, the statement 10.8A is true or snow is white.

10.9 This is not a rejection of logic; logic, of course, stays. E.g. we still do assert that, because snow is white, then the statement 10.8A is true or snow is white. Rather, we must re-think applications of the concept "truth," since so-called evident schemae involving it in fact do not hold. E.g. (again) "snow is white" is true, but "The statement 10.8A is true or snow is white" is not true.

(Philosophical Comments)

Aa Why does philosophy not appear to progress?

Ab One reason is that there is no agreement on the meaning of its terms or its methodology. Unlike mathematics or science, there is no accepted method at arriving at a philosophic truth.

Ac Add to this the fact that the interest in a philosophical assertion is partly determined by its unusualness, and so by how much it diverges from common sense. Thus, when Descartes asserts "God necessarily exists," it is striking partly because it diverges so completely from what we would expect. In fact, on closer examination it even contradicts an evident assertion, namely "no being necessarily exists," and of course there can be no more distinctive idea than one which contradicts a truistic proposition. In order to be heard, one must be interesting, and so there is a natural tendency for a continual generation in philosophy of untrue assertions. On the other hand, a philosopher may assert a proposition which is in fact a truism but claim it is not one, in order for it to have a greater standing, and thus there is a natural tendency for many true assertions to be given invalid justifications. Because there is little agreement on terms or process, it is difficult to reject these claims out-of-hand.

Ad Some of modern philosophy has as its aim the analysis of certain philosophical concepts, e.g. "person" or "good." Such an analysis can in fact be useful if it consists simply of lexicography in an extended sense, i.e. not only compiling all uses, but explaining for as many cases as possible whether the concept applies or not. For instance, how a law applies in a certain set of circumstances may depend on such an analysis. There is one caveat: in the case, e.g., of 'person,' one thinks of bizarre situations and, even when intuition is uncertain, tries to discover whether the concept applies. In doing this, one erroneously pretends that we "know" meanings other than by an intuitive feeling, that is, that we are able to consider them up close and judiciously decide whether they apply to any situation.

Ae But often the analysis is an attempt to find an equivalent reformulation (e.g. listing necessary and sufficient conditions for x to be a person). Now first, there may be no non-trivial equivalent formulation. While we do use a word such as 'person' in a more or less consistent way, there does not seem to be some non-trivial list of properties equivalent to "being a person." And so the search is doomed from the start. Secondly, the only valid arguments for equivalence in the case of undefined, and so intuitively understood, concepts would seem to be either a call to intuition, which would be dubious except for the most trivial reformulations, or an examination of all cases, showing that each case either falls under or does not fall under both the concept and the reformulation. In reality both of these types of arguments are impracticable, and what is used instead (often implicitly) is a justification of, "I'm right until you find a counterexample."

Ba Any statement may be psychologically accepted or dismissed by an individual. It is even, for instance, conceivable that an individual believe he does not exist. From a psychological viewpoint, then, nothing is certain, meaning that no statement must necessarily be accepted by an individual.

Bb There are many ways to justify one's beliefs, among which are:

(1) One is conscious of the fact contained in the belief (that I have an image of a pencil).
(2) It seems immediate from the meanings involved (that a man is a person).
(3) It is supported by a deduction using recognized inferences from recognized axioms (Peano's Axioms imply the Commutative Law of Addition).
(4) One's memory tells one that it is so (that I went to the library today).
(5) One's senses tells one that it is so (that this prencil exists).
(6) It provides the "simplest" explanation to a body of facts where there is as yet no credited evidence to the contrary (that there are two types of electrical charge and that Brutus helped kill Julius Caesar).
(7) One has faith that this is so (that God exists).
There are of course other ways.

Bc A belief held by a deduction depends on the recognized premises and methods of deduction; i.e. if the premises are correct and the methods of deduction valid, the belief is necessarily correct. A belief justified by the sixth method will depend on beliefs justified by the fifth.

Bd But, generally, there is no ordering among the different ways; for instance, the first, second, fifth, and seventh ways are not comparable.

Be Moveover, even if there is an ordering, it is possible that one's beliefs do not respect it. For instance, it is possible that someone believes in the axioms and rules of deduction, but does not believe in the conclusion. The person would be illogical, but it is possible to be illogical.

Bf Because different people may differ as to their acceptance of a method of justification, one person may justifiably believe something which another does not, e.g. that God exists.

Bg One might try to go back one more step and justify various justifications, e.g. providing arguments why one should prefer the justification using one's senses as opposed to justification by faith. However, while I believe that justification by one's senses is preferable to justification by faith, there does not seem to be any argument to support this preference which does not itself have recourse to our senses. Essentially, then, it is a fundamental fact about one's psychology.

Bh About scepticism of the external world there are three important facts: that scepticism is consistent (it is possible to be a sincere sceptic); that non-scepticism is consistent; and that I am not a sceptic. Thus, if I were to try to give a justification for my trust in the external world, it could not be a logically necessary one.

Bi It often seems that modern philosophers fall into the same mistake towards sceptics as Medieval theologians- philosophers did towards atheists: that is, refusing to grant the sceptic and atheist what is obviously theirs, consistency. All one needs to point out against the sceptical position is that it is not logically necessary, and that one does in fact trust that there is an external world.

Bj Often in arguments about the external world or the various ways to justify one's beliefs, one hears mention of "rationality." Now it does seem true that being rational does imply a trust in one's senses. But it should be noted that in the Medieval era being rational implied believing in the existence of God. It is possible that those who are irrational about their senses are correct, and those who are rational are wrong (although of course I do not think so).

Ca My senses present what I see, hear, and touch as exterior to me (my mind).

Cb It is logically possible that I may not be seeing anything. ("There does not exist anything" is a proposition.)

Cc It is logically possible that I may now be seeing incorrectly; I see a bent stick, but it may in fact be straight. It is similarly logically possible that I have always seen incorrectly and always will. Obviously, I do not believe this is the case, but it is possible it is.

Cd Whatever the possibilities, it does seem that I, on the whole, just accept my seeing to be accurate. Only when a seeing instance is contradicted by information otherwise received (perhaps by a seeing instance at another time) am I apt to decide the seeing instance is inaccurate.

Ce Physical theories of sight in no way mean that I am not seeing the pen that I now think I am seeing. For, presumably, I am seeing it. The physical theory, of course, simply gives a causal explanation of how I am able to do this. That it takes photons to strike my eyes in order for me to see something of course does not entail that I am not seeing that thing, nor, for that matter, that I am seeing photons.

Cf But is an object we see as red really red? Of course it is possible it is not (after all, it is always possible we are mistaken). But presumably the object is really red.

Da It is logically possible that there exists a set of generalizations which would allow an individual, given complete data about the state of the present universe, to predict completely all future events. However, even given such a set it would be unlikely that an individual could completely predict the future, because it is unlikely he could have access to enough data about the state of the present universe to permit him to do so.

Db It is likewise logically possible that there exists a set of generalizations which would allow an individual, given certain data, to predict certain events. Indeed, for certain astronomical events such as orbits of the planets, this seems to have come to pass so long as one is not overly ambitious (because one is unlikely to get and store accurate enough data, the predictions cannot reach beyond a limited, albeit very long, time scale). However, certain events, because they are contained in very complex systems, seem unlikely to be predicted in a useful way (such as the weather one year from now, or what a human being will do tomorrow).

Dc So, in the sense of complete prediction by a conscious being, it does not seem that the future will ever be determined.

Dd Usually, though, "determinism" is meant in a less strict way: events are determined if the set of generalizations exists (regardless of whether or not adequate data can be culled to use them practically). As before it is both logically possible that such a set exists or does not (and I have no opinion as to which).

De In any case there could be determinism (in either sense), and so any discussion of free will versus determinism must accept its possibility.

Df On the other hand, it seems that there is choice, that human beings do make choices. Although the possible acts one has to choose from may not be as wide as one might wish (because of history, society, circumstances, etc.), one still does have choice.

Dg Two things that the present individual does not have choice in are: who the present individual is; and what the past individual was and did. (The present person may only effect the future.)

Dh The present person is thus not responsible for who he is, nor what the past person was or did, but only what he does (uncoerced). Of course, if one considers the person persisting over time, then a person is responsible to some extent (varying from person to person) for who he is, as well as any act he does (uncoerced).

Di Suppose for the moment that there is determinism in the stricter sense. Then even under these circumstances there is still choice; that a choice can be predicted does not make it any less a choice.

Dj The I (the present person) still makes choices; it is just that, when I make a choice, the choice is determined, because, side by side, the I is determined. (The past determines the future through the present.)

Dk It is possible that there is predestination, that a being long ago consciously decided what will happen and is now, if needed, carrying out its decision. But even in this case we could still have free choice, so long as the predestination consisted solely of planning the history of the Universe, setting things in motion, and perhaps intervening to change the behaviour of objects other than ourselves. We would not have choice only if the being intervenes and forces us to do what we do not want to; but no one seems to suggest this as an option.

Ea Ethical sentences are often meant assertively (are used to make ethical propositions). "His stealing is bad" asserts something about his stealing, which may be true or untrue.

Eb Ethical propositions describe human acts just like any other proposition which describes human acts. Ethical propositions are descriptive.

Ec In a way then, "ought" can be an "is:" both describe. The propositions "Stealing is bad" and "That boy is small" are both descriptions; they both assert something about their referents. Logically, "ought" and "is" propositions have the same general form (a picture with arrows).

Ed This obviates the 'is-ought' problem. We do not have to derive 'ought' from an 'is.' We already have it; it is meaningful.

Ee But ethical propositions are merely descriptive. Rather than an 'is-ought' problem, there is an 'ought-is' problem, which is to say that from ethical propositions one cannot deduce any significant non-ethical proposition.

Ef "Stealing is bad" ("ought") allows one to say, "One should not steal" ("ought" again); also "One is morally obliged not to steal" ("ought"); hence also, "A good person does not steal" ("ought"); and so, "If a man steals, one has the right to punish him" ("ought" again). We may also deduce from "His stealing is bad" that "He has stolen." But we are unable to deduce that, because stealing is bad, that people are unable to steal, that people do not steal, that people are less likely to steal, that a particular person does not steal, or that a particular person is less likely to steal, the non-ethical propositions we might like to infer. While it seems to be true that most people are less likely to steal than not, and more generally seem less likely to do any act which they consider bad, this is an empirical and not a logical matter. It seems that acts which are bad have usually been translated in the world to acts which are undesirable to commit or acts which we are trained in childhood not to commit, and so most people are less likely to steal. But one could imagine the opposite situation, where people are trained to do bad acts.

Eg Perhaps one might try to deduce a "non-ought" from an "ought" by bringing in the idea of a "perfectly rational being." But the fact is, there need be no one particular way in which a perfectly rational being would act.

Eh Two perfectly rational beings, like two human beings, could act in many different ways, both good and bad, under the same circumstances. War could still exist in a world containing only perfectly rational beings.

Ei Ethics does not bind. That is, it is always possible for me to act as I ought not to.

Ej Ethics places a certain number of acts in a set. But this set does not achieve a special status over other sets of acts by a logical connection, but because of historical and psychological reasons.

Ek Ethical sentences, when they are not meant assertively, seem to be used emotively or meant imperatively. ('You shouldn't do that' sometimes is used to dissuade someone or is meant like "Don't do that.") This triple use seems to result from our learning the descriptive 'should' from the emotive and imperative 'should.' And, it is here whence ethical sentences, even when intended descriptively, derive their peculiar force: from the confusion of different meanings and uses, because each employs the same word.

El Ethical sentences, when used emotively or meant imperatively, are not assertions. They therefore cannot either imply or be implied by ethical propositions. "Don't steal" does not imply and is not implied by "Stealing is bad."

Em If an ethical sentence is used emotively, then a possible proper response is, 'Please don't feel that way.' If it is meant imperatively, a possible proper response is 'Why?' If it is meant descriptively, a possible proper response is 'And so?' And if an ethical sentence is used part emotively, part imperatively, and part descriptively, a possible proper response is, 'I'm confused.'

En Of course, another possible proper response is to agree and carry out the act.

Fa What is the meaning of life? Is there a meaning of life?

Fb Suppose we demand (as it seems we at least partly do) that, for there to be a meaning of life, there must be at least two different possible after-lives, not all equally desirable, and which we enter into depends on how we live our present lives. In this sense absolute death (of the biologist or atheist) would preclude a meaning. On the other hand, Heaven and Hell as conceived in popular lore would then imply there is a meaning to one's life, so long as one desired Heaven more than Hell, or Hell more than Heaven.

Fc On the other hand, suppose we demand (as it seems we at least partly do) that, for there to be a meaning of life, one's life must have more value than non-existence. Then one can, and indeed it seems we do, mean 'value' so that life, or at least lives lived in a certain way, have value. In brief in this sense there does seem to be a meaning of life, since life (or a certain sort of life) has value over non-existence.

Fd But even if there is a meaning of life, and so? Believing there to be a meaning of life (in either sense) does not imply that one will live one's life in a certain way.

Fe It is a fact of my psychology that I would not want to spend any time in Hell; and if I knew that a certain manner of living would assure me of avoiding Hell, then I would live in that way. But I could live otherwise.

Ff It is also a fact of my psychology that I would react differently if the choice were between Heaven and non-existence. If I knew I were going to die in five minutes and that by saying a certain prayer, I would spend the rest of eternity in Heaven, then yes I would say that prayer. But if the period were five years, and the duties more onerous than one prayer, I would probably not do them.

Fg It is a fact of my psychology that I do not spend any time pondering how I might improve the value of my life. It is also a fact of my psychology that I sometimes prefer to act in ways, which incidentally happen to increase the value of my life.

Fh One may commit suicide because one believes there is no meaning to life (in either sense), but there is no logical connection between the two.

Fi Believing that there is no meaning to life (in either sense) no more necessitates suicide than believing there is one.

Fj However, there does seem to be a bit of a causal connection, because for one thing those who believe there is a meaning to life in the first sense often are religious, and religions usually forbid suicide.

Fk If believing there is no after-life causes someone to despair, it is presumably because they regret the idea of no longer living; so it would only be ironic for him to kill himself, and be no longer living even sooner.

Fl Ethics tells us: "One should not commit suicide." And so?

Fm Similarly, one may not care about one's acts, or living generally, because (causally) of a belief that it is eventually going to be the same anyway for everyone. But there is no logical connection between the two. A psychological fact about myself is that I do not make the connection, I do care about my acts, and I want to live.